Discovery of sequential patterns is becoming increasingly useful and valuable in many scientific and commercial applications. Consider for example a Microsoft® Office command sequence. It is valuable information for Microsoft Corporation's developers and support personnel to know how the product is used, such as to know the answer to “What other features are used before or after feature X?” or “What do users do after they visit help?” or “Is feature X easy to find?” (which corresponds to knowing how many clicks are needed in order to execute command X).
However, some sequential data cannot be used as is, but rather needs to be segmented in order to be processed. Precisely segmenting such sequential data into suitable units is a desirable task.